This year, two workshops were held at Schloss Dagstuhl (Wadern, Germany) which involved DDI.
The first, “DDI-CDI: Realising Interoperable Data Services in the Metadata Ecosystem,” took place from 24 to 29 September, and included 25 participants from 21 organizations in 11 countries. The DDI-CDI Working Group was in attendance, supplemented by invited experts. The focus of the workshop was on implementing DDI-CDI, and supporting those who wish to learn more about the standard and implement it. The work was very productive, with the work organized into several different areas:
-
General overview of the specification and the problem space it is meant to address, including how it relates to other standards implemented in a modular way
-
Specific work around the creation of examples and implementation guides, including a “how-to” paper addressing a methodology for creating community specific implementation guides, a worked example mining the metadata embedded in an SPSS file into DDI-CDI’s XML expression, and the development of a prototype tool for use by developers to generate samples, taking a CSV file as input and easily allowing the generation of a DDI-CDI description in JSON-LD (and also other flavors of DDI)
-
A proposal on how non-numeric, non-coded data (e.g. qualitative data) could be described in an integrated fashion with quantitative data sets within the DDI-CDI model
-
Alignment of DDI-CDI with external models which describe the dependencies between variables in cross-domain integration scenarios, based om the RDA’s I-ADOPT standard and the OGC’s Observations’ Measurements standard
-
Syntax representation of the DDI-CDI model, including Python, Typescript, ShEx, SHACL, and JSON Schema.
In each of these areas, the outputs will be incorporated into the work of the DDI-CDI WG moving forward, and will result in the publication of guidance and documentation for users, tools for developers and implementers, and new features for the DDI-CDI model. Also addressed were issues around better supporting implementers through interactions with the DDI Developer’s Group and with a more responsive “beta” release process, and how this could fit into the work of the Alliance overall.
The second Dagstuhl seminar, “Defining a Core Metadata Framework for Cross-Domain Data Sharing and Reuse,” was held from 2-6 October, with 24 participants from 22 organizations in 12 countries. It was focused on the development of the Cross-Domain Interoperability Framework (CDIF), being developed through the WorldFAIR project, in which DDI-CDI is combined with many other domain neutral standards to support easier exchange of metadata and data across disciplinary and infrastructural boundaries. These external standards include Schema.org and DCAT (for discovery and cataloguing), PROV-O (for describing provenance), SKOS, XKOS, and OWL for describing controlled vocabularies and ontologies, and SSSOM for describing mappings. The focus of this work is on the practical implementation of the FAIR principles in cross-domain scenarios, and it promises to provide a strong basis for the adoption of DDI-CDI as part of this suite of standards. Like the first week, this workshop was very productive, and built on the outputs of the week focusing on DDI-CDI which had just taken place.
Thanks to all of the participants and organizers, and especially to the DDI Alliance and CODATA for providing support for this work. As always, Schloss Dagstuhl provides a unique setting for doing focused work, and for making it possible to be incredibly productive.